INDEX

Explanations

mentions or references to the word "Ass" followed by other characters

oai_token-act-pair · gpt-3.5-turbo

variations of the word "assassin" and related terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 spark

-0.73

KT

-0.72

 crushed

-0.72

 grain

-0.71

 fading

-0.70

 finish

-0.69

 magnets

-0.68

 cracks

-0.68

 grit

-0.68

 spir

-0.68

POSITIVE LOGITS

Ass

3.89

Ass

2.12

ass

1.84

ASS

1.83

ASS

1.70

ass

1.43

Asset

1.31

 asses

1.28

 Assist

1.17

Att

1.16

Activations Density 0.010%

mentions or references to the word "Ass" followed by other characters

variations of the word "assassin" and related terms

No Comments

No Known Activations